FedTracker: Furnishing Ownership Verification and Traceability for Federated Learning Model

TL;DR

FedTracker is a novel framework for federated learning that embeds watermarks and fingerprints into models, enabling ownership verification and traceability while maintaining model utility and robustness against attacks.

Contribution

It introduces the first FL protection scheme combining global watermarking and local fingerprinting with continual learning for effective ownership verification and participant traceability.

Findings

Effective ownership verification demonstrated.

High robustness against watermark removal attacks.

Maintains model utility and fidelity.

Abstract

Federated learning (FL) is a distributed machine learning paradigm allowing multiple clients to collaboratively train a global model without sharing their local data. However, FL entails exposing the model to various participants. This poses a risk of unauthorized model distribution or resale by the malicious client, compromising the intellectual property rights of the FL group. To deter such misbehavior, it is essential to establish a mechanism for verifying the ownership of the model and as well tracing its origin to the leaker among the FL participants. In this paper, we present FedTracker, the first FL model protection framework that provides both ownership verification and traceability. FedTracker adopts a bi-level protection scheme consisting of global watermark mechanism and local fingerprint mechanism. The former authenticates the ownership of the global model, while the latter identifies which client the model is derived from. FedTracker leverages Continual Learning (CL) principles to embed the watermark in a way that preserves the utility of the FL model on both primitive task and watermark task. FedTracker also devises a novel metric to better discriminate different fingerprints. Experimental results show FedTracker is effective in ownership verification, traceability, and maintains good fidelity and robustness against various watermark removal attacks.